PopularFX
Home Help Search Login Register
Welcome,Guest. Please login or register.
2022-05-26, 05:19:22
News: If you have a suggestion or need for a new board title, please PM the Admins.
Please remember to keep topics and posts of the FE or casual nature. :)

Pages: 1 2 3 4 5 6 7 8 9 10 11 [12] 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 ... 101
Author Topic: Dally, Shark & Ruslan workbench  (Read 106444 times)

Hero Member
*****

Posts: 853
Itsu,

I tried similar experiment with one of my coils laying around.
I have 20ns pulse generator made with 74ac04. I connected it through 100k variable resistor to one coil end
and adjusted resistor to get forward and reflected pulse amplitudes somewhere close.
Here pictures when second end grounded and left in the air.
Pictures distorted, but I think it is clearly that some kind of reflection is there and it change polarity.
Scope probe on the first coil end.

I guess you will see something similar if you connect scope to the coil.
Of course it has capacitance, but now you have about 50x divider (2.7k vs 50 ohm).

Vasik


---------------------------
The only way of discovering the limits of the possible is to venture a little way past them into the impossible.
   

Group: Professor
Hero Member
*****

Posts: 2746
I made a video of the tiny pulse / bump i see when shorting the far end of the coil.
That is a really tiny pulse!  I don't mind its small amplitude but I do mind its high fall/rise time.  The long coil's terminals hanging outside of the groundplane are not helping.
To do this experiment without the groundplane, you'd need to use a differential picopulser that simultaneously applies a positive pulse and a negative pulse to a break in the coil at its midpoint and converts the reflection from the symmetrical to single-ended form, for visualization on the single-ended grounded scope.  I don't think that any ferrite-based balun can handle picopulses.
Anyway, that ~100ns reflection (~200ns round trip) most likely corresponds to the 11.7MHz frequency.

If you make a VNA S11 or VSWR measurement, the best results would be obtained by connecting* the VNA to a break in the wire at the midpoint of the coil, ...or winding a second identical coil and connecting* the VNA between these 2 coils.
If you do not want to break the wire or wind a 2nd coil then you are left with driving the coil with a low-turn loop, but this loop will not influence the coil at only one point like an electric connection ...and this loop has its own RF characteristics that hopefully can be calibrated out.

*  with an SMA panel socket of course ;)
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3433
Itsu,

I tried similar experiment with one of my coils laying around.
I have 20ns pulse generator made with 74ac04. I connected it through 100k variable resistor to one coil end
and adjusted resistor to get forward and reflected pulse amplitudes somewhere close.
Here pictures when second end grounded and left in the air.
Pictures distorted, but I think it is clearly that some kind of reflection is there and it change polarity.
Scope probe on the first coil end.

I guess you will see something similar if you connect scope to the coil.
Of course it has capacitance, but now you have about 50x divider (2.7k vs 50 ohm).

Vasik

Vasik, 

that worked nicely, at least how i understood that, i have added another probe (yellow) to the start of the coil (groundlead not used) and it shows the flipping over when the far end of the coil is connected to the copper trip, see video https://youtu.be/4v3nhiSpCOM

Itsu
   

Hero Member
*****

Posts: 853
Vasik, 

that worked nicely

Itsu

O0


---------------------------
The only way of discovering the limits of the possible is to venture a little way past them into the impossible.
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3433
That is a really tiny pulse!  I don't mind its small amplitude but I do mind its high fall/rise time.  The long coil's terminals hanging outside of the groundplane are not helping.
To do this experiment without the groundplane, you'd need to use a differential picopulser that simultaneously applies a positive pulse and a negative pulse to a break in the coil at its midpoint and converts the reflection from the symmetrical to single-ended form, for visualization on the single-ended grounded scope.  I don't think that any ferrite-based balun can handle picopulses.
Anyway, that ~100ns reflection (~200ns round trip) most likely corresponds to the 11.7MHz frequency.

If you make a VNA S11 or VSWR measurement, the best results would be obtained by connecting* the VNA to a break in the wire at the midpoint of the coil, ...or winding a second identical coil and connecting* the VNA between these 2 coils.
If you do not want to break the wire or wind a 2nd coil then you are left with driving the coil with a low-turn loop, but this loop will not influence the coil at only one point like an electric connection ...and this loop has its own RF characteristics that hopefully can be calibrated out.

*  with an SMA panel socket of course ;)


I think the TDR measurement are not really usefull in finding the Wave resonance of this coil.

So i used my nanoVNA to see if using its VSWR function will show anything usefull.
As i do not want to break up my coil in the middle (yet), i opted for a midpoint exciting of this coil using a wire loop.

Using severall loops ranging from ½ turn loop to now 3 tuns loop, i think this last one performs well.

First screenshot is a 10KHz to 15MHz sweep range showing a peak around 6MHz.
Second screenshot is an extended range to 50MHz showing severall peaks.

I think it shows (VSWR plot) that exciting this coil in the middle so as a ½ wave dipole, its wave resonance frequency is 6MHz, meaning if excited as a ¼ wave coil (what we need) its wave resonance frequency is 3Mhz as per the calculator.

Itsu   

   

Group: Administrator
Hero Member
*****

Posts: 3179
It's not as complicated as it may seem...
The reflection was always there Itsu, but was being snuffed out by the pulser. That's why I suggested you remove the pulser after the pulse is sent.

The 2k7 resistor may or may not be allowing maximum power into the coil (I'm not yet convinced), but I think you would see an even bigger reflection if you used no resistor, but sensed the reflection with the pulser disconnected. You might try a diode in series (A-K) with your pulser as an auto-disconnect. Of course you will need to select the correct polarity of reflection (positive, so open-ended) for it to work.

Anyway, it appears now that your probe is on the correct side of the resistor, you have enough signal to determine what you need.
   

Group: Professor
Hero Member
*****

Posts: 2746
So i used my nanoVNA to see if using its VSWR function will show anything usefull.
It does show the classic dipole response, which is very useful for determining its electric length, but I think there might be a problem with the OSL calibration of your VNA.

I did not pay attention to it before, but now I will look into it.

First of all, the excitation loop must be connected to the VNA during all three stages of the calibration: open, short and load.  :o
Because the DUT is located inside the excitation loop, only the conditions inside that loop must change during the three calibration stages  ???

When the exciation loop is empty (i.e the Stalker's coil is absent from it), that must be the OPEN condition for the VNA. In that condition the calibrated VSWR should approach infinity and │S11│ - unity.  Is that true after your calibration ?
You may ask why ?  A: Because there is no current flowing at the midpoint of the Stalker coil when that coil is entirely removed from the loop (obviously).  The excitation loop does have current flowing in it, but you are not interested in it because you are trying to measure the current flowing in the midpoint of the Stalker coil - not in the excitation loop.

The SHORT condition would be equivalent to entirely filling the excitation loop with a copper cylinder*, because that simulates the lowest impedance that this loop can achieve.  In that condition the calibrated VSWR should approach infinity and │S11│ - unity, as well.

Finally, the LOAD condition would be equivalent to filling the excitation loop with a slotted copper pipe* that has a 50Ω resistor** across that slot.

Sorry to complicate it so much, but these are the consequences of using an excitation loop and calibrating it properly.  I only hope it makes sense.


* or at least a copper pipe or disk or ring
** The impedance of an ideally matched dipole is 73Ω+43j so a 73Ω resistor would seem better, but Stalker's coil is closer to a helical antenna, which has a different impedance when matched.  I do not know what that impedance is for a low pitch helix antenna like this. It would be interesting to determine that matching impedance by seeking the lowest VSWR when fed with 1/2 wavelength resonance frequency.
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3433
The reflection was always there Itsu, but was being snuffed out by the pulser. That's why I suggested you remove the pulser after the pulse is sent.

The 2k7 resistor may or may not be allowing maximum power into the coil (I'm not yet convinced), but I think you would see an even bigger reflection if you used no resistor, but sensed the reflection with the pulser disconnected. You might try a diode in series (A-K) with your pulser as an auto-disconnect. Of course you will need to select the correct polarity of reflection (positive, so open-ended) for it to work.

Anyway, it appears now that your probe is on the correct side of the resistor, you have enough signal to determine what you need.


Right, the flipping signal does signal something, not sure what as the numbers don't add up for me.

The now yellow negative going pulse when i short the coil to the copper strip (1m) can be seen after 189.3ns (delta yellow cursors).

With speed of light in air/vacuum being around 0.299m/ns (29.9cm/ns) this reflected signal is equivalent to 0.299 x 189.3 = 56.6m

Taken the guestimated velocity factor of 0.66 (66%), it calculates to 37.35m which is so very close to the real 37.5m length of the coils wire.

If the signal still travels forth and back it would mean we should divide by 2 meaning 18.67m

Is this (almost exact) 50% reduction then due to the fact that this is a coil, not a straight wire?

Itsu
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3433
It does show the classic dipole response, which is very useful for determining its electric length, but I think there might be a problem with the OSL calibration of your VNA.

I did not pay attention to it before, but now I will look into it.

First of all, the excitation loop must be connected to the VNA during all three stages of the calibration: open, short and load.  :o
Because the DUT is located inside the excitation loop, only the conditions inside that loop must change during the three calibration stages  ???

When the exciation loop is empty (i.e the Stalker's coil is absent from it), that must be the OPEN condition for the VNA. In that condition the calibrated VSWR should approach infinity and │S11│ - unity.  Is that true after your calibration ?
You may ask why ?  A: Because there is no current flowing at the midpoint of the Stalker coil when that coil is entirely removed from the loop (obviously).  The excitation loop does have current flowing in it, but you are not interested in it because you are trying to measure the current flowing in the midpoint of the Stalker coil - not in the excitation loop.

The SHORT condition would be equivalent to entirely filling the excitation loop with a copper cylinder*, because that simulates the lowest impedance that this loop can achieve.  In that condition the calibrated VSWR should approach infinity and │S11│ - unity, as well.

Finally, the LOAD condition would be equivalent to filling the excitation loop with a slotted copper pipe* that has a 50Ω resistor** across that slot.

Sorry to complicate it so much, but these are the consequences of using an excitation loop and calibrating it properly.  I only hope it makes sense.


* or at least a copper pipe or disk or ring
** The impedance of an ideally matched dipole is 73Ω+43j so a 73Ω resistor would seem better, but Stalker's coil is closer to a helical antenna, which has a different impedance when matched.  I do not know what that impedance is for a low pitch helix antenna like this. It would be interesting to determine that matching impedance by seeking the lowest VSWR when fed with 1/2 wavelength resonance frequency.

thanks, that makes sense, even to me.

I will redo the calibration using those steps / procedures and see what is going to happen, afterall; the proof of the pudding is in the eating.

Itsu
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3433

I did a quick and dirty new calibration using some copper foil as tubing, see picture, but the result is kind of chaotic for now, see screenshot


Will redo this later today.

Itsu
   

Group: Professor
Hero Member
*****

Posts: 2746
I did a quick and dirty new calibration using some copper foil as tubing, see picture, but the result is kind of chaotic for now, see screenshot
The tubing's walls are very thin this way. Maybe you could make them out of the same blue wire connected as identical loops (several of them can be stacked).

Is the diameter of the calibration tubing the same as the Stalker coil's ?
The 50Ω resistor in slotted tubing is most likely wrong, try to experiment with different values...I find it hard to calculate the impedance of a matched double helix dipole when it has low turn-pitch like yours. Maybe Smudge could help?
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3433

Ok,  the foil loops fit inside the 3 turn coil similar as the Stalker coil, so yes, about the same diameter.

I will try with the same blue wire to stack severall single turn loops.

Thanks,  Itsu
   

Group: Administrator
Hero Member
*****

Posts: 3179
It's not as complicated as it may seem...

Right, the flipping signal does signal something, not sure what as the numbers don't add up for me.

The now yellow negative going pulse when i short the coil to the copper strip (1m) can be seen after 189.3ns (delta yellow cursors).

With speed of light in air/vacuum being around 0.299m/ns (29.9cm/ns) this reflected signal is equivalent to 0.299 x 189.3 = 56.6m

Taken the guestimated velocity factor of 0.66 (66%), it calculates to 37.35m which is so very close to the real 37.5m length of the coils wire.

If the signal still travels forth and back it would mean we should divide by 2 meaning 18.67m

Is this (almost exact) 50% reduction then due to the fact that this is a coil, not a straight wire?

Itsu

Do you really have 37.5 meters of wire on that coil?

A long solenoid when considered as a TL has got to be complicated. If it has a tapered characteristic impedance, it therefore may also have a tapered VF.

A coil of given straight wire certainly shortens its "length", but in what way? A coil has self-capacitance* plus capacitance to the environment (Ce). A straight length of wire only has Ce. In both scenarios, the determination VF is complicated by the fact that the "return" signal experiences two dielectrics, one of the wire jacket, and one of free space. A difficult to predict TL over all I would think.

* Does the self-capacitance "bypass" the windings, shortening the electrical length?
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3433

Yes, exact 37.5m including the connection leads.

Its not tapered, so the velocity factor should be equal across the length.

Thanks,  Itsu
   

Group: Administrator
Hero Member
*****

Posts: 3179
It's not as complicated as it may seem...
Maybe the pulse is not strictly traveling to the opposite end via the wire?

With the inter-winding capacitance, the coil is bypassed from end-to-end. This small capacitance would in theory cause the applied pulse to arrive at the coil end before it does via the wire/TL. At that point, it may even be reapplied to that end of the coil.

If a pulse appears to be traversing a TL twice as fast as it should, then either the pulse is traveling at 2c*VF, or the reflection is occurring at the midway point of the coil, or the TL is half the expected length.
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3433

Poynt,

good questions, but i have no answers.

I tried to pulse my old Grenade (3.4mm wire 47m long) see picture, but this gives no clear reflection like the long coil did, probably due to the increased capacitance etc.

Guess i will stop investigating into pulse reflection etc. as it serves no real purpose here.




I did use my nanoVNA to measure the VSWR from a 144MHz dipole antenna with some 5m of coax.
Calibration as normal with open, short and 50 Ohm load and it shows the low SWR around the target frequency.

So measuring the long coil has its challenges, but also here i guess i will go to the next step being to coil it up to a Grenade.


Itsu
   

Group: Moderator
Sr. Member
*****

Posts: 335
Enjoy your trek through life but leave no tracks
Yes this is what we find, Lets just deviate a moment, I was saying earlier i have an old 37.5 coil  the best amplitude
at what must tune up was 5mhz or just over a few khz, and not much on the harmonics side.

I have been trying to re tune the coil using a selection of caps across the 37.5Mtr winding, peeks appear to be
best in 5khz steps ie 15khz 20 khz up tp 50khz, my capacitor rang is some what limited but i can get 5 sets of harmonics
3 of high amplitude 100khz apart and 2 others at half frequency in-between also 100khz apart but attenuated.

If any one is interested.
   

Group: Professor
Hero Member
*****

Posts: 2746
I have been trying to re tune the coil using a selection of caps across the 37.5Mtr winding
Caps come with their own gotchas.
https://youtu.be/BcJ6UdDx1vg?t=1466
   

Group: Moderator
Sr. Member
*****

Posts: 335
Enjoy your trek through life but leave no tracks
I'm sure they do, nothing is perfect in this life, but then again wold you expect the resonant peakes to be where
we would expect them to be where the coil harmonics are ? And then 5mhz is a little to high for a TL494 to generate
and then there is 1/2 wave and 1/4 wave frequencys to wory about.

There is also the grenade winding on page 3 the maths doesnt add up, in that the first 2 windings 56 and 55 taking up
1/2 the 37.5  winding ie 18.75 that bit sounds ok

then we have layer 3 and 4 , 26 and 25 winds  and then layer 5 and 6  with 16 and 14 turns 

That doesn't come to 18.75 mtrs   ;D ;D some thing isn't right there ? or is it me ?  ;D ;D
   

Group: Professor
Hero Member
*****

Posts: 2746
I'm sure they do, nothing is perfect in this life, but then again wold you expect the resonant peakes to be where
we would expect them to be where the coil harmonics are ?
Did you ever perform an experiment like the one described here ?
   

Group: Moderator
Sr. Member
*****

Posts: 335
Enjoy your trek through life but leave no tracks
Did you ever perform an experiment like the one described here ?
Well back in my youth when i worked for a government owned telecom company, They sent us back to Technical college, from what i can remember Audio is a longdetudinal wave (different frequencies they had to RF modulate AM) all you need to do here to simulate the phenomena is
audio modulate the carrier as in ham radio.


Suppose I have a tesla coil running at 100khz and another one running at 140khz  what freq would you expect if you used that idea on the output ?

SIL
   
Hero Member
*****

Posts: 2179
Hey AG

Quote
Suppose I have a tesla coil running at 100khz and another one running at 140khz  what freq would you expect if you used that idea on the output ?

On a similar thought...
Many people tried to build Viktor Schaubergers "Repulsine" wave disk as two identical sine waves, one sine wave on top of the other. However, in fact Viktor offset the lower sine wave disk inward to the center so that the flow in between the disks was restricted/accelerated on the inlet side and expanded/decelerated on the outlet side of the wave form. This is the equivalent of using two wave forms to create a third effect independent of the two forms which created it like a beat note.

If this physical effect Viktor produced was also true of electrical effects then I would expect to see a differential charge density or potential gradient across the beat note. Not unlike the concept Nikola Tesla called "wheels within wheels" or an electrical charge carrier based planetary gear set.

I have found Nikola Tesla's thoughts on this matter to be true in my experiments because in essence electrical effects are physical effects. In effect we are compressing, expanding and moving charge carriers not unlike a physical fluid and it follows similar rules.

Regards
AC



« Last Edit: 2021-04-28, 21:54:28 by Allcanadian »


---------------------------
Comprehend and Copy Nature... Viktor Schauberger

“Progress is impossible without change, and those who cannot change their minds cannot change anything.” George Bernard Shaw
   

Group: Professor
Hero Member
*****

Posts: 2746
Suppose I have a tesla coil running at 100khz and another one running at 140khz  what freq would you expect if you used that idea on the output ?
40kHz, 100kHz, 140kHz, 280kHz  ...and possibly multiples of these frequencies.
   

Group: Moderator
Sr. Member
*****

Posts: 335
Enjoy your trek through life but leave no tracks
So then, don't we have the device Akula took to Berlin. And it also describes Adrian Gustoves account of his device.

Well lets have another look at this aspect over on OU user Color published this wave form and again Ruslan
refers to it, I feel this so called dancing wave again Ruslan and Color refer, lets have a look.

1 in the vertical sense we have am modulation
2 in the horizontal sense we have a more complicated wobble of phases

Would any one like to offer any circuits or useful comments on how this exact effect could me simulated ?

SIL
« Last Edit: 2021-04-29, 10:00:59 by AlienGrey »
   

Group: Elite Experimentalist
Hero Member
*****

Posts: 3433
I made out of my long single coil (37.5m of 4.2mm / 2.5mm² diameter wire) a Grenade according to the version 2 layout given by Vasik here:
https://www.overunityresearch.com/index.php?topic=3926.msg90863#msg90863                     see picture 1.


For calculations of number of turns etc. i used the calculator (excel) as given by Vasik here:
https://www.overunityresearch.com/index.php?topic=3926.msg90731#msg90731                     see picture 2


The result can be seen in picture 4.

The inductance measurement (143uH @ 100KHz) is spot on for the calculations made from the long coil (143uH) here:
https://www.overunityresearch.com/index.php?topic=3926.msg91005#msg91005                    see picture 3


I will do some LC and wave resonance measurements with this Grenade.

Itsu
« Last Edit: 2021-04-29, 14:15:12 by Itsu »
   
Pages: 1 2 3 4 5 6 7 8 9 10 11 [12] 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 ... 101
« previous next »


 

Home Help Search Login Register
Theme © PopularFX | Based on PFX Ideas! | Scripts from iScript4u 2022-05-26, 05:19:22